Fast, Scalable Training
Train AI models with high-performance infrastructure that scales effortlessly. Simplified setup ensures rapid deployment and accelerated results.
Cloud GPU powered
compute and Kubernetes
compute and Kubernetes
High-performance NVIDIA® GPUs for machine learning and generative AI.
Unbeatable pricing from just $0.69 per GPU/h
The latest NVIDIA GPUs including B200 Blackwell, H100, H200, A100 and L40S
Support for compute and Kubernetes
Trusted by businesses of all sizes worldwide.
Instant, affordable access to GPU clusters
Real-Time Inference
Achieve instant, accurate AI inference with low-latency architecture, enabling seamless decision-making and optimized user experiences.
Accelerated Data Processing
Accelerate complex computations and data transformations with powerful computing capabilities. Process raw data efficiently for faster insights.
Built for Performance and Scale
Our platform is built for high performance and scalability, delivering the speed and capacity to handle any workload effortlessly. You can even combine our GPUs with Kubernetes for even more power and flexibility, allowing for seamless scaling without the complexity.
UK Sovereignty Guaranteed
Your data remains securely within UK borders at all times, ensuring it is never transferred internationally or subjected to foreign laws. We guarantee full compliance with UK regulations, offering complete data sovereignty, security, and peace of mind.
Sustainability at Our Core
With our decarbonized GPU offerings, we are committed to sustainability while driving innovation. Our environmentally conscious infrastructure reduces your carbon footprint and supports your growth with cutting-edge technology.
Maximum GPU power at the lowest possible price
Industry-leading pricing to make AI innovation accessible to everyone
| Model | Status | From price | On-demand price |
|---|---|---|---|
| A100 40GB | In Stock now | $0.69 Per GPU/h | $1.09 Per GPU/h |
| A100 80GB | In Stock now | $1.39 Per GPU/h | $1.79 Per GPU/h |
| L40S 40GB | In Stock now | $0.89 Per GPU/h | $1.29 Per GPU/h |
| H100 PCIe | In Stock now | $1.99 Per GPU/h | $2.49 Per GPU/h |
| H100 SXM | In Stock now | $2.49 Per GPU/h | $2.99 Per GPU/h |
| H200 SXM | In Stock now | $2.99 Per GPU/h | $3.49 Per GPU/h |
| B200 | In Stock now | $3.49 Per GPU/h | Price on request |
For full details on pricing and the savings you can make see our pricing page here
Accelerate your AI and ML projects
From single GPU Compute Instances to scalable GPU Kubernetes cluster
GPU Kubernetes clusters
The adaptability of Kubernetes meets GPU firepower. Designed for vast datasets and intricate models. Offering easy scalability and ROI by supporting multiple GPUs efficiently.
GPU Compute instances
Compute instances powered by our NVIDIA GPUs. Perfect for individual workloads and experiments, machine learning model testing, and graphics rendering.
Trusted by global businesses
The first thing that drew us to Civo was the expertise they demonstrated in selecting the right GPUs for our needs. Their knowledge made all the difference.
Matt Butcher
Chief Executive Officer, Fermyon
For our PoC, Civo impressed us by setting up a Slack Connect channel for direct access to their SRE team. Their ledge of cloud-native tech and GPUs gave us real confidence in their support.
Oliver Pinson-Roxburgh
Chief Executive Officer, Defense.com
Unlike other industry players, Civo doesn't bury hidden charges or surprise you with complex egress fees - their pricing is straightforward, and that makes all the difference for businesses trying to scale efficiently.
James Faure
Chief Executive Officer, Clairo.ai
With Civo, deployment speeds and efficiency have significantly improved. Their GPU expertise has been invaluable, boosting our service delivery and enhancing client conversion and retention.
Anuraag Gutgutia
Co-Founder, True Foundry
Discover Civo's AI/ML learning resources
Secure your GPU today
Frequently Asked Questions
We offer a range of NVIDIA GPUs, including B200 Blackwell, H100, H200, A100, and L40S. You can choose the model that best fits your workload and budget, from high-memory options for large-scale training to more cost-effective instances for smaller experiments. Check our Pricing Table for the latest availability and hourly rates.
- Choosing the right GPU among the NVIDIA A100, L40S, H100, H200, and B200 depends on your specific needs and use cases. Consider the balance of AI, graphics, and HPC requirements in your projects—focusing on memory, processing power, and the nature of your workloads (AI training/inference, graphics rendering, or HPC tasks).
- - NVIDIA A100: Delivering unmatched AI and graphics performance, the A100 is the pinnacle of innovation for demanding workloads in AI, data analytics, and scientific computing. It features cutting-edge technology that excels in multi-tenant environments or large-scale data centers.
- - NVIDIA L40S: Ideal if you need a balanced blend of AI and graphics performance. Powered by the Ada Lovelace architecture with 48 GB of GDDR6 memory, the L40S handles intricate AI computations and advanced graphics processing—perfect for 3D rendering, large language model training, and other dual-purpose tasks.
- - NVIDIA H100: Best suited for intensive AI training and inference tasks. It offers up to 80GB of HBM2e memory and is equipped with fourth-generation Tensor Cores, making it a superior choice for building large AI models such as chatbots and recommendation engines. With robust security features and scalability for multi-GPU systems, the H100 is an enterprise workhorse for data center operations.
- - NVIDIA H200: Building on the H100’s capabilities, the H200 provides additional memory and faster networking, ensuring high throughput and speed. This makes it a top pick for large-scale AI training and inference where enhanced data transfer rates and bigger memory capacity are crucial—for instance, in generative AI and complex AI inference models.
- - NVIDIA B200 (Blackwell Architecture): Designed on the next-generation Blackwell architecture, the B200 is NVIDIA’s latest leap forward for extreme AI and HPC demands. With significantly boosted memory bandwidth, improved parallel computing capabilities, and advanced multi-GPU scalability, the B200 is optimized for next-generation generative AI, large-scale simulation, and the most demanding research workloads.
- Key Considerations When Choosing a GPU:
- 1. Memory Requirements: Larger models or HPC tasks may necessitate high memory (e.g., H100, H200, B200).
- 2. Type of Workload: For blended AI/graphics workloads, look at L40S.
- 3. Scalability & Networking: If you plan multi-GPU clusters, verify your GPU supports robust interconnect features (H100, H200, B200).
- 4. Budget & Pricing: Balance upfront costs versus ongoing performance needs; e.g., A100 offers high performance at competitive rates, while B200 caters to specialized, premium-level workloads. For a deeper technical comparison of these GPUs, check out our blog post: A100 vs. L40S vs. H100 vs. H200GH Superchips By aligning your AI, graphics, and HPC requirements with the specs and cost of each GPU model, you can select the option that maximizes performance without overspending. If you need additional guidance, feel free to contact our sales team for a tailored recommendation.
- It depends on your workflow:
- - GPU Compute Instances are ideal for quick experiments, single-model deployments, or workloads that don’t require complex orchestration.
- - GPU Kubernetes is perfect if you want to leverage container orchestration, auto-scaling, and microservices architectures. It also enables you to manage multiple containerized workloads efficiently on the same cluster.
At Civo, we are dedicated to providing climate-friendly GPU products. Our sustainable cloud solution, powered by 100% renewable energy, runs on Deep Green's carbon-neutral data center. We effectively capture 90% of the heat generated by our servers, repurposing it for community-based projects, thus reducing our carbon footprint and contributing to a sustainable future.
As the exclusive cloud partner of Deep Green, we offer state-of-the-art GPU-powered servers that align with environmental responsibility. We believe in making sustainability accessible; hence, we've maintained competitive pricing across all regions to encourage more businesses to adopt eco-friendly computing solutions.
Civo offers plug-and-play integration and easy configuration, ensuring users familiar with our existing instance services feel right at home. With our setup, it's effortless to scale from startup to hyperscale without any vendor lock-in.
- Yes, alongside our flexible on-demand hourly pricing, we offer discounted rates for longer-term commitments to help you save as you scale. Here’s a breakdown of our options
- - On-demand
- - 6-month commitment
- - 12-month commitment
- - 24-month commitment
- - 36-month commitment
While we aim to support a wide range of use cases, including large deployments, there may be initial limits for new accounts. These can be increased upon request and a quick review, ensuring we can accommodate your scaling needs while maintaining performance for all users.
Civo is certified with CE and CE+ marks, ensuring compliance with European safety, health, and environmental protection standards. We are also ISO 27001 certified, signifying our commitment to rigorous information security management practices. Additionally, we adhere to SOC 2 standards, focusing on trust principles like security, availability, and confidentiality, ensuring that our services align with recognized best practices for protecting your data and operations.
Trusted compliance services from a certified provider
